The goal of this proposal is to localize human epileptic networks by characterizing their electrophysiological activity over a wide range of spatiotemporal scales. Decades of clinical intracranial EEG (IEEG) using restricted spatial (centimeter scale) and temporal (~0.5-100 Hz) bandwidth, based more on tradition than modern neuroscience, have frustrated epileptologists looking for discrete, resectable "electrographic lesions" during evaluation for epilepsy surgery. Similarly, recent efforts to apply direct brain stimulation to abort seizures after they are sufficiently established to be detected on standard clinical macroelectrodes have, so far, met with only partial success. We hypothesize that enhancing the spatial and temporal resolution of clinical intracranial EEG can improve the efficacy of epilepsy surgery and responsive brain stimulation to control seizures. Human epileptic networks produce pathological activity that ranges from seizures and spikes, generated by cubic centimeters of brain tissue, to high frequency oscillations that occur on sub-millimeter dimensions. Recent evidence suggests that important components of these signals are found at frequencies not detected by standard clinical IEEG. Using simultaneous IEEG recordings from microwire arrays and clinical macroelectrodes, our group has begun to characterize two potential signatures of epileptogenic brain, high frequency oscillations and "micro-seizures," that are outside the resolution of conventional clinical IEEG. In this application, we propose analysis of continuous, high-resolution, wide- bandwidth IEEG recorded simultaneously from microwire arrays and clinical macroelectrodes in order to localize human epileptic networks. We will correlate our findings with surgical outcome, prospectively, in a cohort of patients undergoing evaluation for epilepsy surgery. This work builds upon our established effort in Translational Neuroengineering melding state of the art epilepsy care with cutting-edge research. PUBLIC HEALTH RELEVANCE The neuronal networks of human epileptic brain are multiscale;extending from cellular assemblies organized on the scale of cortical columns (~300 - 600 <m) to large-scale networks organized over lobar structures. These pathological networks generate oscillations over a wide range of frequency (0.01 - 1000 Hz) and spatial scales not probed by clinical EEG. Our laboratory and others have identified pathological network oscillations occurring outside the range of clinical IEEG that appear to be signatures of the epileptogenic zone.